The RNA Chaperone Hfq Impacts Growth, Metabolism and Production of Virulence Factors in Yersinia enterocolitica Tamara Kakoschke 1 , Sara Kakoschke 1 , Giuseppe Magistro 1 , So ¨ ren Schubert 1 , Marc Borath 2 , Ju ¨ rgen Heesemann 1 , Ombeline Rossier 1 * 1 Max von Pettenkofer Institute for Hygiene and Medical Microbiology, Ludwig Maximilians University, Munich, Germany, 2 Protein Analysis Unit, Adolf-Butenandt Institute, Ludwig Maximilians University, Munich, Germany Abstract To adapt to changes in environmental conditions, bacteria regulate their gene expression at the transcriptional but also at the post-transcriptional level, e.g. by small RNAs (sRNAs) which modulate mRNA stability and translation. The conserved RNA chaperone Hfq mediates the interaction of many sRNAs with their target mRNAs, thereby playing a global role in fine- tuning protein production. In this study, we investigated the significance of Hfq for the enteropathogen Yersina enterocolitica serotype O:8. Hfq facilitated optimal growth in complex and minimal media. Our comparative protein analysis of parental and hfq-negative strains suggested that Hfq promotes lipid metabolism and transport, cell redox homeostasis, mRNA translation and ATP synthesis, and negatively affects carbon and nitrogen metabolism, transport of siderophore and peptides and tRNA synthesis. Accordingly, biochemical tests indicated that Hfq represses ornithine decarboxylase activity, indole production and utilization of glucose, mannitol, inositol and 1,2-propanediol. Moreover, Hfq repressed production of the siderophore yersiniabactin and its outer membrane receptor FyuA. In contrast, hfq mutants exhibited reduced urease production. Finally, strains lacking hfq were more susceptible to acidic pH and oxidative stress. Unlike previous reports in other Gram-negative bacteria, Hfq was dispensable for type III secretion encoded by the virulence plasmid. Using a chromosomally encoded FLAG-tagged Hfq, we observed increased production of Hfq-FLAG in late exponential and stationary phases. Overall, Hfq has a profound effect on metabolism, resistance to stress and modulates the production of two virulence factors in Y. enterocolitica, namely urease and yersiniabactin. Citation: Kakoschke T, Kakoschke S, Magistro G, Schubert S, Borath M, et al. (2014) The RNA Chaperone Hfq Impacts Growth, Metabolism and Production of Virulence Factors in Yersinia enterocolitica. PLoS ONE 9(1): e86113. doi:10.1371/journal.pone.0086113 Editor: Michael Hensel, University of Osnabrueck, Germany Received July 17, 2013; Accepted December 5, 2013; Published January 15, 2014 Copyright: ß 2014 Kakoschke et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported in part by grants from the Wissenschaftlichen Herausgeberkollegium der Mu ¨ nchener Medizinischen Wochenschrift and from the Bayerischen Gleichstellungsfo ¨ rderung to O.R. T.K. received a fellowship from the Fo ¨ rderprogramm fu ¨ r Forschung und Lehre of the Ludwig-Maximilians- University Medical School. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: [email protected]Introduction The genus Yersinia includes three human pathogenic species, namely Y. pestis, the agent of plague and two enteropathogenic species, Y. pseudotuberculosis and Y. enterocolitica. We study the Gram-negative bacterium Y. enterocolitica as a model for an extracellular enteropathogen. Upon ingestion of contaminated food or water, Y. enterocolitica is able to invade the intestinal submucosa and preferentially multiplies extracellularly in Peyer’s patches and mesenteric lymph nodes [1,2]. Y. enterocolitica virulence factors include proteins important for early stages of infection, such as urease, a multisubunit metalloenzyme which facilitates survival to stomach acidity [3,4] or the outer membrane adhesin called invasin which promotes transcytosis across the epithelial barrier [5]. Two other major virulence factors, which are essential for later stages of infection, are encoded by the virulence plasmid pYV: the outer membrane adhesin YadA and the type III secretion system Ysc (Ysc-T3SS). The Ysc-T3SS is a complex machinery that translocates at least 6 anti-host effector proteins into the host cell (YopH, YopM, YopO, YopT, YopP and YopE), where they collectively inhibit phagocytosis and dampen the inflammatory response [6,7]. In addition to the pathogenicity factors mentioned above, strains of Y. enterocolitica biogroup 1B, which are highly virulent in a mouse model of infection, carry a so- called high pathogenicity island (HPI). The HPI encodes proteins involved in production and import of the siderophore yersinia- bactin [8]. These proteins include the transcriptional activator YbtA, the biosynthetic enzymes Irp1-Irp5 and Irp9, the inner membrane ABC transporters Irp6 and Irp7, and the yersiniabactin receptor FyuA, which is localized in the outer membrane [8–15]. FyuA also confers sensitivity to the bacteriocin pesticin [11]. Importantly, yersiniabactin production and utilization is an essential virulence trait for Y. enterocolitica in mouse infection [10,11,16]. Genes involved in pathogenicity of enteropathogenic Yersinia ssp. are regulated by environmental factors such as temperature, ionic strength, pH and host cell contact. For example, under in vitro conditions, urease and invasin are most highly expressed at 27uC, the optimal growth temperature [17,18]. In contrast, pYV plasmid genes encoding the Yop proteins, Ysc-T3SS and the adhesin YadA are upregulated at 37uC, the temperature of the mammalian host PLOS ONE | www.plosone.org 1 January 2014 | Volume 9 | Issue 1 | e86113
17
Embed
The RNA Chaperone Hfq Impacts Growth, Metabolism and ... · The RNA Chaperone Hfq Impacts Growth, Metabolism and Production of Virulence Factors in Yersinia enterocolitica Tamara
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
The RNA Chaperone Hfq Impacts Growth, Metabolismand Production of Virulence Factors in YersiniaenterocoliticaTamara Kakoschke1, Sara Kakoschke1, Giuseppe Magistro1, Soren Schubert1, Marc Borath2,
Jurgen Heesemann1, Ombeline Rossier1*
1 Max von Pettenkofer Institute for Hygiene and Medical Microbiology, Ludwig Maximilians University, Munich, Germany, 2 Protein Analysis Unit, Adolf-Butenandt
Institute, Ludwig Maximilians University, Munich, Germany
Abstract
To adapt to changes in environmental conditions, bacteria regulate their gene expression at the transcriptional but also atthe post-transcriptional level, e.g. by small RNAs (sRNAs) which modulate mRNA stability and translation. The conservedRNA chaperone Hfq mediates the interaction of many sRNAs with their target mRNAs, thereby playing a global role in fine-tuning protein production. In this study, we investigated the significance of Hfq for the enteropathogen Yersinaenterocolitica serotype O:8. Hfq facilitated optimal growth in complex and minimal media. Our comparative protein analysisof parental and hfq-negative strains suggested that Hfq promotes lipid metabolism and transport, cell redox homeostasis,mRNA translation and ATP synthesis, and negatively affects carbon and nitrogen metabolism, transport of siderophore andpeptides and tRNA synthesis. Accordingly, biochemical tests indicated that Hfq represses ornithine decarboxylase activity,indole production and utilization of glucose, mannitol, inositol and 1,2-propanediol. Moreover, Hfq repressed production ofthe siderophore yersiniabactin and its outer membrane receptor FyuA. In contrast, hfq mutants exhibited reduced ureaseproduction. Finally, strains lacking hfq were more susceptible to acidic pH and oxidative stress. Unlike previous reports inother Gram-negative bacteria, Hfq was dispensable for type III secretion encoded by the virulence plasmid. Using achromosomally encoded FLAG-tagged Hfq, we observed increased production of Hfq-FLAG in late exponential andstationary phases. Overall, Hfq has a profound effect on metabolism, resistance to stress and modulates the production oftwo virulence factors in Y. enterocolitica, namely urease and yersiniabactin.
Citation: Kakoschke T, Kakoschke S, Magistro G, Schubert S, Borath M, et al. (2014) The RNA Chaperone Hfq Impacts Growth, Metabolism and Production ofVirulence Factors in Yersinia enterocolitica. PLoS ONE 9(1): e86113. doi:10.1371/journal.pone.0086113
Editor: Michael Hensel, University of Osnabrueck, Germany
Received July 17, 2013; Accepted December 5, 2013; Published January 15, 2014
Copyright: � 2014 Kakoschke et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported in part by grants from the Wissenschaftlichen Herausgeberkollegium der Munchener Medizinischen Wochenschrift and fromthe Bayerischen Gleichstellungsforderung to O.R. T.K. received a fellowship from the Forderprogramm fur Forschung und Lehre of the Ludwig-Maximilians-University Medical School. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
SOR17 JB580v derivative with a deletion of hfq marked with a KmR cassette This study
SOR35 JB580v derivative with an unmarked chromosomal fusion of hfq with sequencesencoding the 3xFLAG epitope
This study
8081-U-GB R2M+ derivative of clinical isolate 8081, yeuA::Km, urease-negative [4]
E. coli
DH5a [98]
CC118lpir [99]
S. enterica serotype Typhimurium
WR1542 reporter strain WR1330 (fepA::Tn10dTc, iroN::pGP704, cir::MudJ) carrying plasmidpACYC5.2L with genes promoting the import of yersiniabactin (fyuA, irp6-8), theirtranscriptional activator (ybtA) and a fusion of the fyuA promoter to luciferase
and PduG) (Table 4), a metabolic activity believed to promote
adaptation of S. Typhimurium and Listeria monocytogenes to
particular niches in host tissues [51]. The last spot found to be
more abundant in the hfq mutant is a putative periplasmic binding
protein encoded by gene ye2751, which flanks the pdu region, and,
unlike the pdu genes, is conserved in Y. pseudotuberculosis and Y.
pestis. Based on conserved domain CD06302, YE2751 could be
involved in the transport of pentose or hexose sugars.
Figure 1. Growth of Y. enterocolitica strains in BHI (A, B) and LB (C,D). (A and B) Bacteria were grown in BHI at 27uC (A) and 37uC (B): parentalstrains WA-314 (black diamonds) and JB580v (black squares), hfq-negative strains SOR3 (white diamonds), SOR4 (white triangles) and SOR17 (whitesquares). (C and D) Growth in LB of complemented strains at 27uC (C) and 37uC (D): JB580v(pACYC184) (black squares and straight line, parental strainharbouring the control plasmid), JB580v(pAhfq) (white square and dotted line, parental strain with complementing plasmid), SOR17(pACYC184)(black triangle and black line, hfq-negative strain with control plasmid), and SOR17(pAhfq) (white triangle and dotted line, complemented hfq strain).Full complementation of the growth defect of strains SOR3 and SOR4 was also observed after introduction of plasmid pAhfq (data not shown).Results are the mean and standard deviation (error bars) of two cultures and are representative of at least two independent experiments.doi:10.1371/journal.pone.0086113.g001
Yersinia enterocolitica RNA Chaperone Hfq
PLOS ONE | www.plosone.org 6 January 2014 | Volume 9 | Issue 1 | e86113
Taken together, the results of the 2-DE analysis suggest that Hfq
impacts metabolism, surface proteins and stress responses of Y.
enterocolitica.
Influence of hfq on carbohydrate metabolismAs the next step in our analysis, we explored the influence of
Hfq on carbohydrate metabolism. First, we performed biochem-
ical tests using the API-20E kit. Using a pH indicator dye, this kit
detects the release of organic acids upon bacterial growth in the
presence of different carbohydrates. Amygdalin was the only
carbohydrate for which all hfq–negative strains exhibited reduced
medium acidification compared to the corresponding parental
strains (data not shown). Interestingly, growth in the presence of
inositol led to increased medium acidification for hfq mutants
SOR3 and SOR4 compared to the parental strain WA-314 (data
not shown), similarly to what has been described in Y. enterocolitica
O:9 [26]. Strains lacking hfq exhibited typical acidification of the
media after growth for 24 h at 27 uC in the API-20E wells
containing glucose, mannitol, sorbitol, sucrose and arabinose.
However, we wondered if the slowed growth of hfq mutants could
mask differences between parental and mutant strains. Bacterial
suspensions inoculated to the API-20E strips are very dilute and an
increase in acidification might not be detectable because parental
and mutant bacteria are at different stages of growth. A hint that
this might be the case came from the following observation: upon
growth on Yersinia selective agar (CIN agar) for two days at 27 uC,
all strains produced colonies with the typical dark pink bull’s eye
pattern, indicative of mannitol utilization (mannitol is the only
carbohydrate in CIN agar). However, we noticed that all the hfq-
negative strains also produced a strong pink halo surrounding
areas of heavy bacterial growth. Indeed, when we spotted bacterial
suspensions on CIN agar, we observed that after two days of
Table 3 hfq-dependent changes in protein abundance found by 2-DE analysis upon growth in LB for 5 h at 37uC.
Spot # RegulationaMW(kDa) YE # Gene name Protein description GO Biological processb
Soluble proteins
1302 – 32 YE3057 ybbN putative thioredoxin
2802 + 71 YE3090 htpG heat shock protein 90 protein folding; response to stress
a, +: protein more abundant in hfq-negative strain; –: protein less abundant in hfq-negative strain; b, Gene ontology biological function used in the GenoList database(http://genodb.pasteur.fr); c, N/A: not applicable.doi:10.1371/journal.pone.0086113.t003
Yersinia enterocolitica RNA Chaperone Hfq
PLOS ONE | www.plosone.org 7 January 2014 | Volume 9 | Issue 1 | e86113
incubation at 27uC, spots of hfq mutants were surrounded by a
sharp dark pink halo that intensified over the next 2 days (Fig. 3A).
The wild type also produced a halo but only one or two days later,
suggesting that acidification of the agar medium was quicker for
the hfq-negative strains. Complementation of this phenotype was
achieved with plasmid pAhfq (Fig. 3A and data not shown). To
confirm that the halo appearance was independent of the dye used
to monitor acidification, we also spotted bacterial cultures on an
agar medium containing mannitol and phenol red (instead of
neutral red): indeed, strain SOR4 produced a yellow halo that
appeared earlier and was stronger than the one produced by wild-
type WA-314 (data not shown). We also grew bacteria on
MacConkey agar supplemented with different sugars. Since Y.
enterocolitica does not utilize lactose, all strains grown on
unsupplemented MacConkey produced yellow colonies, whereas
plates supplemented with mannitol, glucose or sucrose gave rise to
red/pink colonies (data not shown). When bacterial suspensions
were spotted on MacConkey agar with mannitol or glucose, the
pink halos were stronger for hfq mutants SOR4 and SOR17
compared to their parental strains WA-314 and JB580v. On
MacConkey agar containing sucrose, we could only observe a very
faint pink halo around all the spots with no noticeable differences
between parental strains and mutants (data not shown). Finally, we
also used MacConkey agar containing 1,2-PD and vitamin B12,
an essential co-factor for the Pdu enzyme complex. Similar to
medium containing sucrose, acidification around bacterial spots of
the wild types was very faint (Fig. 3A). Because hfq mutants grew
slightly more slowly on this medium, we could not easily compare
them to their parental strains (Fig. 3A). However, we noticed that
expressing additional copies of hfq in the parental strains (from
plasmid pAhfq) led to a reduction in the pink color of spots or
colonies (Fig. 3A and B), evoking a decrease in 1,2-PD utilization.
Overall, our results suggests that Hfq represses the catabolism of
mannitol, glucose, inositol and 1,2-PD in Y. enterocolitica.
Figure 2. 2-DE analysis of total soluble (A) and total membrane (B) proteins stained with Coomassie blue. Bacteria were grown intriplicate at 37uC for 5 h. One representative gel per strain is shown. Proteins were separated in 2-DE gels (for all gels: pH range 3–10, molecularweight (MW) range 15–150 kDa). Highlighted spots were identified by mass spectrometry (see Table 3). MW marker size is indicated in kDa.doi:10.1371/journal.pone.0086113.g002
Yersinia enterocolitica RNA Chaperone Hfq
PLOS ONE | www.plosone.org 8 January 2014 | Volume 9 | Issue 1 | e86113
Influence of hfq on nitrogen metabolismNext we examined the influence of hfq on nitrogen metabolism.
Since our 2-DE proteome analysis suggested that tryptophanase
was more abundant in a hfq mutant, we used Kovacs’ reagent to
detect the production of indole, the by-product of tryptophanase
activity. Upon growth in LB at 27uC, hfq mutants produced more
indole than the parental strains, a phenotype that was comple-
mented for bacteria carrying plasmid pAhfq (Fig. 4A and B).
Therefore, together with results from 2-DE, our analysis indicates
that Hfq represses the production of tryptophanase.
Using the API-20E strips, we observed that the ornithine
decarboxylase activity was also markedly increased for all strains
lacking hfq when bacteria were grown at 27 uC (Fig. 4C),
suggesting that polyamine synthesis is also modulated by hfq.
Finally, we also noted that urease activity was decreased for all hfq
mutants compared to their corresponding parental strains in the
API-20E strips (data not shown). To further assess the influence of
hfq on the production of urease, we performed immunoblotting
using a rabbit polyclonal antibody specific for the 19-kDa UreB
subunit [17]. Bacteria were grown overnight at 27uC, conditions
described for maximal urease production [3,4]. Figure 5 shows that
the urease production is reduced in hfq-negative strains relative to
wild types, although the reduction observed in the WA-314
derivatives was more modest than in strain JB580v (50% and
80% reduction, respectively). Complementation was observed after
Table 4 hfq-dependent changes in protein abundance found by 2-DE analysis upon growth in LB for 16 h at 27uC.
Regulationa MW (kDa) YE # gene name Protein description GO biological functionb
+ 80 YE1771 fcuA ferrichrome receptor protein siderophore transport
+ 50 YE0650 tnaA tryptophanase tryptophan catabolic process
+ 33 YE2731 pduD/pddB putative propanediol utilization protein:dehydratase, medium subunit
+ 32 YE2729 pduB putative propanediol utilization protein response to external stimulus
+ 10 YE2728 pduA putative propanediol utilization protein
a, +: protein more abundant in hfq-negative strain; b, Gene ontology biological function used in the GenoList database (http://genodb.pasteur.fr).doi:10.1371/journal.pone.0086113.t004
Figure 3. Influence of hfq on carbohydrate metabolism. (A) Bacteria were spotted on CIN agar (top row) and MacConkey agar supplementedwith vitamin B12 and 1,2-PD (bottom row). Plates were incubated at 27uC for three (top) or two days (bottom). (B) Bacteria were grown onMacConkey agar supplemented with vitamin B12 and 1,2-PD at 27uC for two days.doi:10.1371/journal.pone.0086113.g003
Yersinia enterocolitica RNA Chaperone Hfq
PLOS ONE | www.plosone.org 9 January 2014 | Volume 9 | Issue 1 | e86113
introduction of pAhfq in the mutant strains (Fig. 5B compare lane 1
and 2). Thus, Hfq enhances the production of urease, a known
virulence/fitness factor of Y. enterocolitica.
Role of hfq in susceptibility to acidic, oxidative andantibiotic stress
Since urease is known to contribute to resistance to acidic pH,
we tested whether a hfq-negative strain would be more susceptible
to acid stress using a survival assay. As shown in Fig. 6A, both hfq-
negative strains SOR4 and SOR17 exhibited a reduced survival at
pH 4.0 compared to their parental strains. Mirroring its more
pronounced decrease in urease production, strain SOR17 was
more susceptible to acidic stress than strain SOR4 (6% compared
to 26% respectively). Using plasmid pAhfq, we observed
complementation of the survival defect of strain SOR17 (Fig.
6A). Hence, in Y. enterocolitica, Hfq promotes resistance to acidic
stress.
As a next step in our study, we analyzed bacterial susceptibility
to additional stress challenges, i.e. oxidative and antibiotic stress.
Both hfq mutants SOR4 and SOR17 were more susceptible to
killing by hydrogen perdoxide that parental strains, with again an
Figure 4. Influence of hfq on indole production and ornithinedecarboxylase activity. (A and B) The concentration of indolepresent in culture supernatants was determined after growth in LB at27uC. (A) Bacteria were grown for four hours at 27uC. Data representmean and standard deviation of at least three independent experimentseach performed with triplicate independent cultures. (B) Complemen-tation analysis. Bacterial cultures were grown for 16 h at 27uC, sincestrains carrying plasmids were delayed in their indole production.Because of the variability of indole concentration produced by parentalstrains carrying plasmids (between 0.07 and 1.5 mM in four indepen-dent experiments), results were expressed relative to the indoleproduced by the parental strain JB580v carrying the control vector(-which was set at 100%). Data represent mean and standard deviation offour independent experiments each performed with at least triplicateindependent cultures. Significance was calculated with Student‘sunpaired t-test (*P,0.05; **P,0.01; ***P,0.001). (C) Ornithine decar-boxylase activity detected using the API-20E strip. All wells are positive(negative wells remain yellow), but wells inoculated with hfq-negativestrains turn red, whereas those inoculated with parental strains aremore orange.doi:10.1371/journal.pone.0086113.g004
Figure 5. Immunodetection of the 19-kDa urease beta subunitin total protein extracts of Y. enterocolitica. The relative signal foreach band compared to wild type (which was set to 100%) is indicated.Upper panel shows the immunoblot, bottom panel shows part of theCoomassie blue-stained gel used as loading control. (A) Loading was asfollows: 1, WA-314; 2, SOR4; 3, SOR3; 4, urease-negative control strain8081-U-GB; 5, JB580v, and 6, SOR17. (B) Complementation analysis.Loading was as follows: 1, SOR4(pACYC184ts); 2, SOR4(pAhfq); 3, WA-314(pAhfq); 4, WA-314(pACYC184ts); 5, WA-314; and 6, 8081-U-GB. Inanother experiment, we also observed restoration of the production ofUreB in the hfq-negative strain SOR17 carrying pAhfq (data not shown).doi:10.1371/journal.pone.0086113.g005
Yersinia enterocolitica RNA Chaperone Hfq
PLOS ONE | www.plosone.org 10 January 2014 | Volume 9 | Issue 1 | e86113
even stronger phenotype in strain SOR17 (Fig. 6B).Introduction of
the complementation plasmid pAhfq into SOR17 increased the
strain’s resistance to H2O2 (Fig. 6B). Therefore, as described for
other bacteria and Yersinia species [25,27,28], Hfq promotes
survival of Y. enterocolitica in the presence of oxidative stress.
Finally we determined the minimal inhibitory concentration
(MIC) of several antibiotics for WA-314 and SOR4: no significant
differences in MIC were observed between strains for all
antibiotics tested, i.e. ampicillin, oxacillin, gentamicin and
trimethoprim/sulfamethoxazole (data not shown). Therefore, lack
of hfq does not lead to general increase in sensitivity to antibiotics.
Influence of hfq on production of the siderophorereceptor FyuA
Among the OMPs whose production was increased in the hfq
mutant, the 2-DE proteomic analysis identified FyuA (Table 3),
which is an essential virulence factor in Y. enterocolitica biotype 1B
strains [11]. FyuA functions as the receptor for the siderophore
yersiniabactin but also for the bacteriocin pesticin [11]. To
confirm the influence of Hfq on FyuA production under
conditions where iron is not depleted, we performed a pesticin
susceptibility assay using a disk diffusion assay (Table 5). As
observed previously [11], a strain lacking fyuA is resistant to killing
by pesticin, as denoted by the absence of growth inhibition even at
the highest concentration of pesticin (Table 5). In contrast, the hfq-
negative strain SOR4 was more susceptible to pesticin compared
to the parental strain WA-314: SOR4 showed an increase in both
the size of the growth inhibition zone and in the minimum dilution
factor required to observe growth inhibition (MID), a phenotype
that was complemented by expressing hfq from plasmid pAhfq
(Table 5). In the course of the complementation experiment, we
also noted that pAhfq rendered the wild-type strain WA-314
completely resistant to pesticin (Table 5). Using this assay, we
observed some strain differences: strain JB580v appeared more
susceptible than strain WA-314 to pesticin. Lack of hfq renders
JB580v only slightly more susceptible to the bacteriocin with a
modest 2-fold increase in the MID (Table 5). In summary, Hfq
appears to repress susceptibility to pesticin, which is likely to reflect
its influence on the production of FyuA.
Next we tested whether Hfq also played a role in FyuA
production under low-iron conditions (to alleviate Fur repression).
Bacteria were grown for 24 h at 37uC in LB supplemented with
the ferrous iron chelator DIP (LBD), and then FyuA was detected
by immunoblotting (Fig. 7A). The outer membrane receptor was
more abundant in hfq-negative strains than in parental strains (ca.
30–50% increase). Most strikingly, increased production of Hfq
from plasmid pAhfq led to an 80% reduction in FyuA in the wild
type strains. Taken together, our results indicate that Hfq inhibits
the production of FyuA.
Role of hfq in siderophore productionSince the transcriptional regulator YbtA regulates expression of
fyuA as well as the genes involved in yersiniabactin biosynthesis, we
next tested whether Hfq played a role in yersiniabactin produc-
tion. Using a reporter strain which contains a yersiniabactin-
Figure 6. Influence of hfq on bacterial survival to acidic andoxidative stress. (A) Bacterial survival to exposure to pH 4.0 for 90min. (B) Bacterial survival to exposure to 1 mM H2O2 for 90 min. Resultsare expressed as % survival relative to bacteria incubated in PBS pH 7.5and are the mean and standard deviation of at least three experimentsperformed with three separate cultures. Complementation assayscorrespond to two independent experiments performed with at leastthree separate cultures. Significance was calculated with Student‘sunpaired t-test (*P,0.05; **P,0.01; ***P,0.001). Bacterial strains areWA-314 and its hfq-negative derivative SOR4, JB580v and its hfq-negative derivative SOR17.doi:10.1371/journal.pone.0086113.g006
Table 5 Pesticin sensitivity assaya.
Strains GenotypeHalo diameterb
(cm) MIDc
WA-314 wt 1.0 2
SOR4 hfq 1.3 12
WA fyuA fyuA 0 ,1
WA-314(pAhfq) wt (hfq+) 0 ,1
SOR4(pACYC184) hfq (vector) 1.5 16
SOR4(pAhfq) hfq (hfq+) 0.8 1
JB580v wt 1.2 16
SOR17 hfq 1.2 32
a, all strains were tested in duplicate at least twice and a representativeexperiment is shown; b, size of growth inhibition obtained with undilutedpesticin preparate; c, MID: minimum inhibitory dilution factor for pesticinpreparate to inhibit bacterial growth.doi:10.1371/journal.pone.0086113.t005
Yersinia enterocolitica RNA Chaperone Hfq
PLOS ONE | www.plosone.org 11 January 2014 | Volume 9 | Issue 1 | e86113
responsive promoter fused to luciferase, we were able to detect
yersiniabactin released into the supernatants of bacteria grown in
LBD. As shown in Fig. 7B, the hfq-negative strains SOR4 and
SOR17 produced nearly twice as much siderophore as their
respective parental strains. Complementation was achieved by
expressing hfq from pAhfq (Fig. 7B). Our data indicate that Hfq
represses production of the siderophore yersiniabactin in
Y. enterocolitica.
Role of hfq in type III secretionGiven the essential role of Ysc-T3SS for the pathogenicity of
Y. enterocolitica [6], we next investigated the role of Hfq in protein
secretion. Following growth under inducing conditions, i.e. at
37uC in Ca2+-depleted media, Yop effector proteins secreted into
the supernatant were analyzed by SDS-PAGE and Coomassie
blue staining. All mutants were tested on at least four different
occasions in either low-Ca2+ LB or low-Ca2+ BHI, and we
observed no major differences between the profile of Yop proteins
in the supernatants of hfq mutants and those of the parental strains
(Fig. 8). Immunoblotting also confirmed that YopB, YopD, LcrV,
YopP, YopE, YopM, YopH and YopQ were secreted in
comparable amounts by parental strains and hfq mutants (Fig. 8
and data not shown). Moreover, the amount of Yops detected in
cell lysates was also not influenced by the absence of Hfq (Fig. 8).
These results are in contrast with those obtained with
Y. pseudotuberculosis, where Hfq promotes the production of Yops
[28], and thus point to some difference in Hfq-mediated regulation
of virulence factors between the two enteropathogenic Yersinia
species. When grown at 37uC in LB or BHI with intrinsic Ca2+
levels for 1.5 h (conditions allowing some Yop production but not
secretion), strains JB580v and SOR17 also produced comparable
amounts of cell-associated YopH (data not shown).
Figure 7. Role of hfq in production of yersiniabactin and its receptor FyuA. (A) Immunodetection of FyuA in strains grown for 24 h in LBsupplemented with DIP (LBD). Loading was as follows: 1, WA fyuA; 2, WA-314; 3, SOR4; 4, JB580v; 5, SOR17; 6, WA-314(pACYC184ts); 7, WA-314(pAhfq); 8, SOR4(pACYC184ts); and 9, SOR4(pAhfq). Upper panel shows the immunoblot. The relative signal for each band compared to wild type(which was set to 100%) is indicated. Bottom panel shows part of Coomassie blue-stained gel used as loading control. (B) Reporter assay measuringyersiniabactin production. Following growth for 24 h in LBD at 37uC, bacterial culture supernatants were harvested. They were applied to a reporterstrain which expresses luciferase in response to yersiniabactin. Luciferase activity was determined after incubation of the reporter strain for 24 h at37uC. Results are the mean and standard deviation of duplicate cultures each assessed in triplicate. Significance was calculated with Student‘sunpaired t-test (**P,0.01; ***P,0.001). Similar results were obtained in three independent experiments.doi:10.1371/journal.pone.0086113.g007
Figure 8. Analysis of Yop proteins secreted by Y. enterocolitica.Proteins secreted into the supernatant (SN, lanes 1-4, 9–10) andproteins from total bacterial cell extracts (Cells, lanes 5–8, 11–12) wereanalyzed by Coomassie blue staining (upper panel) and by immuno-blotting using antibodies specific for YopB, YopD, LcrV, YopE and YopP.Loading was as follows: molecular weight markers (in kDa); 1 and 5,parental strain WA-314; 2 and 6, hfq mutant SOR3; 3 and 7, hfq mutantSOR4; 4 and 8, TTSS-defective lcrD mutant strain WA-314(pYV-515); 9and 11, parental strain JB580v; 10 and 12, hfq-negative strain SOR17.doi:10.1371/journal.pone.0086113.g008
Yersinia enterocolitica RNA Chaperone Hfq
PLOS ONE | www.plosone.org 12 January 2014 | Volume 9 | Issue 1 | e86113
Production of Hfq in Y. enterocolitica serotype O:8To facilitate Hfq detection in Y. enterocolitica strains, we tagged
the chromosomal hfq gene with sequences encoding the FLAG
epitope to generate strains SOR33 and SOR35. The fusion
appears to be functional as both strains exhibited normal growth
in LB in contrast to hfq mutants (data not shown). Production of
Hfq-Flag from plasmid pAhfqFlag was also able to complement
the growth of hfq mutants (data not shown), confirming that the
fusion protein is functional. We next analyzed the time course of
production of Hfq-Flag in Y. enterocolitica grown at 27uC and 37uCin LB in four independent experiments. We observed an increase
in the amount of Hfq-Flag in late exponential phase and stationary
phase compared to early exponential phase (ranging from 300 to
800% upon growth at 27uC and 300 to 1200% at 37uC) (Fig. 9
and data not shown). Therefore, Hfq-Flag accumulates to higher
levels towards the end of exponential phase and beyond.
Discussion
In this study, we have phenotypically characterized hfq mutants
in two strains of different lineages of Y. enterocolitica serotype O:8,
strains WA-314 and JB580v. Loss of hfq led to the same
phenotypes in both strains, indicating that Hfq plays a conserved
role in Y. enterocolitica serotype O:8. We made several observations
indicating that the metabolism of Y. enterocolitica is profoundly
influenced by the RNA chaperone Hfq, encompassing the
metabolism of carbohydrates, nitrogen, iron and fatty acids, as
well as ATP synthesis. In the first step of our analysis we observed
that all hfq mutants exhibited a slowed growth and entered
stationary phase at a lower OD, a phenotype often (but not always)
associated with loss of hfq in other bacteria, including pathogens
[25,30,50]. In other pathogenic Yersinia spp., inactivation of hfq
was reported to affect growth to different degrees. Y. pestis strains
lacking hfq were most altered in growth [27,29], especially at 37uC,
whereas Y. pseudotuberculosis hfq mutants had only minor growth
defects [28,29]. Therefore, Y. enterocolitica appears to have an
intermediate phenotype. Moreover, in contrast to Y. pseudotuber-
culosis [28], lack of Hfq does not affect Yop production and
secretion by the Ysc-T3SS in Y. enterocolitica serotype O:8 strains.
Taken together, our results suggest that Hfq and potential Hfq-
associated sRNAs could affect metabolism and regulation of
pathogenicity factors differently among the pathogenic Yersinia
species.
Because of the central role of Hfq in post-transcriptional
regulation, deletion of the hfq gene results in pleiotropic
phenotypes in many bacteria. In Salmonella enterica sv. Typhimur-
ium, a mutation in hfq leads to differential expression of 20% of all
genes [52,53], whereas in Y. pestis ca. 6% of all genes were affected
[27]. Such a broad regulatory effect may be explained by the
impact of Hfq on the regulation of transcriptional regulators, such
as sigma factors [25,54], but also by the high number of mRNAs
that interact with Hfq. Indeed, up to 15% of S. Typhimurium
mRNAs are thought to directly interact with Hfq [53]. The Hfq
hexamer is believed to bind mRNAs on the proximal side and
sRNAs on its distal side [22]. Two studies have defined a
consensus for mRNA sequences bound to Hfq. The first one
analyzed the quaternary structure of Hfq bound to RNA and
defined a region with four or five (ARN) triplet repeats where R is
a purine nucleotide and N any nucleotide [55]. The second study
identified a consensus by genomic SELEX: AAYAAYAA, where Y
represents pyrimidines (C or U) [56]. An inspection of the genome
of Y. enterocolitica strain 8081 shows that both consensus can be
found in 38 annotated mRNAs within 40 nucleotides of the
ribosome binding site (preliminary results), suggesting that Hfq
might interact directly with these mRNAs and yet unknown
sRNAs to regulate their stability and/or translation.
Carbohydrate metabolismIn this study we observed that Hfq represses carbohydrate
metabolism in Y. enterocolitica. Enzymes associated with glycolysis
(PykF) and the pentose phosphate pathway (TktA and TalB) were
more abundant in the cellular extracts of the hfq mutant.
Moreover, we observed increased media acidification upon growth
in API-20E wells containing inositol, and upon growth on agar
media containing glucose or mannitol. Interestingly, in Y. pestis, a
strain mutated in hfq shows an increase in transcripts encoding
PykF and MtlK, a putative mannitol transporter, suggesting that
Hfq also represses glycolysis and carbohydrate transport in this
pathogenic species [27]. In addition, we observed that 1,2-PD
utilization (Pdu) is repressed by Hfq: (1) in our 2-DE analysis
performed with strains grown overnight at 27uC, five Pdu proteins
(PduA-D and PduG) were more abundant in the hfq-negative
strain, (2) when grown in the presence of 1,2-PD, overexpression
of Hfq in wild-type strains led to a decrease in media acidification.
1,2-PD is a by-product of fucose and rhamnose metabolism that is
found in the gut. Although 1,2-PD utilization has so far not been
investigated in Y. enterocolitica, it was mainly studied in S.
Typhymurium where it is believed to promote growth in vivo as
well as intracellular multiplication in macrophages [57,58],
suggesting metabolic adaptation to niches relevant to pathogenesis.
Figure 9. Immunodetection of Hfq-Flag in total protein extractsof Y. enterocolitica. Time course of expression of Hfq-Flag duringgrowth in LB at 37uC (A) and at 27uC (B). (A) Bacterial extracts of WARS
derivatives (odd-numbered lanes) or JB580v derivatives (even-num-bered lanes) were prepared after 2, 4, 6, 8 and 24 h of growth at 37uC.Loading was as follows: extracts from SOR33 in lanes 1, 3, 5, 7 and 9;SOR35 in lanes 2, 4, 6, 8 and 10; parental strain WARS in lanes 11 and 13;and parental strain JB580v in lanes 12 and 14. The upper band indicatedby an asterisk is a background band also present in parental strainsWARS and JB580v (lanes 11–14) and it was used as loading control. (B)Bacterial extracts of JB580v derivatives were prepared after growth for2, 3, 4, 6, 8 and 12 h at 27uC. Loading was as follows: extracts fromSOR35 in lanes 1–6 and parental strain JB580v in lane 7.doi:10.1371/journal.pone.0086113.g009
Yersinia enterocolitica RNA Chaperone Hfq
PLOS ONE | www.plosone.org 13 January 2014 | Volume 9 | Issue 1 | e86113
In S. Typhimuriun, pdu gene expression is controlled by the
transcriptional activator PocR [59] and several global regulators,
including Hfq [52,60]. In Yersinia spp., pdu genes are restricted to a
subset of species, and are notably absent from the genomes of
Y. pestis and Y. pseudotuberculosis [61,62], suggestive of adaptation to
different niches.
Nitrogen metabolismBesides carbohydrate metabolism, several proteins involved in
nitrogen metabolism were also influenced by Hfq: i.e. OppA,
ornithine decarboxylase and tryptophanase. Our 2-DE analysis
revealed that Y. enterocolitica Hfq represses production of OppA, a
27. Geng J, Song Y, Yang L, Feng Y, Qiu Y, et al. (2009) Involvement of the post-
transcriptional regulator Hfq in Yersinia pestis virulence. PLoS One 4: e6213.
28. Schiano CA, Bellows LE, Lathem WW (2010) The small RNA chaperone Hfq is
required for the virulence of Yersinia pseudotuberculosis. Infect Immun 78: 2034–
2044.
29. Bai G, Golubov A, Smith EA, McDonough KA (2010) The importance of the
small RNA chaperone Hfq for growth of epidemic Yersinia pestis, but not Yersinia
pseudotuberculosis, with implications for plague biology. J Bacteriol 192: 4239–
4245.
Yersinia enterocolitica RNA Chaperone Hfq
PLOS ONE | www.plosone.org 15 January 2014 | Volume 9 | Issue 1 | e86113
30. Sittka A, Pfeiffer V, Tedin K, Vogel J (2007) The RNA chaperone Hfq isessential for the virulence of Salmonella typhimurium. Mol Microbiol 63: 193–217.
31. Shakhnovich EA, Davis BM, Waldor MK (2009) Hfq negatively regulates typeIII secretion in EHEC and several other pathogens. Mol Microbiol 74: 347–363.
32. Koo JT, Alleyne TM, Schiano CA, Jafari N, Lathem WW (2011) Globaldiscovery of small RNAs in Yersinia pseudotuberculosis identifies Yersinia-specific
small, noncoding RNAs required for virulence. Proc Natl Acad Sci U S A 108:
E709–717.
33. Beauregard A, Smith EA, Petrone BL, Singh N, Karch C, et al. (2013)
Identification and characterization of small RNAs in Yersinia pestis. RNA Biol 10.
34. Hoiseth SK, Stocker BA (1981) Aromatic-dependent Salmonella typhimurium are
non-virulent and effective as live vaccines. Nature 291: 238–239.
35. Blattner FR, Plunkett G, 3rd, Bloch CA, Perna NT, Burland V, et al. (1997) The
complete genome sequence of Escherichia coli K-12. Science 277: 1453–1462.
36. Gerhardt P, editor (1994) Methods for general and molecular bacteriology.
Washington, DC: American Society for Microbiology. 791 p.
37. Chant EL, Summers DK (2007) Indole signalling contributes to the stable
maintenance of Escherichia coli multicopy plasmids. Mol Microbiol 63: 35–43.
38. Datsenko KA, Wanner BL (2000) One-step inactivation of chromosomal genes
in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci U S A 97: 6640–6645.
39. Rose RE (1988) The nucleotide sequence of pACYC177. Nucleic Acids Res 16:356.
40. Uzzau S, Figueroa-Bossi N, Rubino S, Bossi L (2001) Epitope tagging ofchromosomal genes in Salmonella. Proc Natl Acad Sci U S A 98: 15264–15269.
41. Sasse J, Gallagher SR (2009) Staining proteins in gels. Curr Protoc Mol BiolChapter 10: Unit 10 16.
42. Roggenkamp A, Schubert S, Jacobi CA, Heesemann J (1995) Dissection of the
Yersinia enterocolitica virulence plasmid pYVO8 into an operating unit andvirulence gene modules. FEMS Microbiol Lett 134: 69–73.
43. Trulzsch K, Roggenkamp A, Aepfelbacher M, Wilharm G, Ruckdeschel K, etal. (2003) Analysis of chaperone-dependent Yop secretion/translocation and
effector function using a mini-virulence plasmid of Yersinia enterocolitica. Int J MedMicrobiol 293: 167–177.
strain-dependent cytosolic accumulation and post-translational secretion.
Microbiology 148: 1457–1465.
46. Heesemann J, Gross U, Schmidt N, Laufs R (1986) Immunochemical analysis of
plasmid-encoded proteins released by enteropathogenic Yersinia sp. grown incalcium-deficient media. Infect Immun 54: 561–567.
47. Tsui HC, Feng G, Winkler ME (1996) Transcription of the mutL repair, miaA
tRNA modification, hfq pleiotropic regulator, and hflA region protease genes of
Escherichia coli K-12 from clustered Esigma32-specific promoters during heatshock. J Bacteriol 178: 5719–5731.
48. Saken E, Roggenkamp A, Aleksic S, Heesemann J (1994) Characterisation ofpathogenic Yersinia enterocolitica serogroups by pulsed-field gel electrophoresis of
genomic NotI restriction fragments. J Med Microbiol 41: 329–338.
Differential patterns of acquired virulence genes distinguish Salmonella strains.Proc Natl Acad Sci U S A 95: 4641–4645.
59. Bobik TA, Ailion M, Roth JR (1992) A single regulatory gene integrates controlof vitamin B12 synthesis and propanediol degradation. J Bacteriol 174: 2253–
2266.
60. Figueroa-Bossi N, Lemire S, Maloriol D, Balbontin R, Casadesus J, et al. (2006)
Loss of Hfq activates the sigmaE-dependent envelope stress response inSalmonella enterica. Mol Microbiol 62: 838–852.
61. Thomson NR, Howard S, Wren BW, Holden MT, Crossman L, et al. (2006)
The complete genome sequence and comparative genome analysis of the high
inhibits translation initiation of Escherichia coli hfq by binding to a single site
overlapping the Shine-Dalgarno sequence. J Bacteriol 189: 5472–5481.
87. Sharma AK, Payne SM (2006) Induction of expression of hfq by DksA is essential
for Shigella flexneri virulence. Mol Microbiol 62: 469–479.
88. Ramachandran VK, Shearer N, Jacob JJ, Sharma CM, Thompson A (2012)
The architecture and ppGpp-dependent expression of the primary transcrip-
tome of Salmonella Typhimurium during invasion gene expression. BMC Genomics
13: 25.
89. Mildiner-Earley S, Miller VL (2006) Characterization of a novel porin involved
in systemic Yersinia enterocolitica infection. Infect Immun 74: 4361–4365.
Yersinia enterocolitica RNA Chaperone Hfq
PLOS ONE | www.plosone.org 16 January 2014 | Volume 9 | Issue 1 | e86113
90. Reines M, Llobet E, Dahlstrom KM, Perez-Gutierrez C, Llompart CM, et al.
(2012) Deciphering the acylation pattern of Yersinia enterocolitica lipid A. PLoSPathog 8: e1002978.
91. Cathelyn JS, Ellison DW, Hinchliffe SJ, Wren BW, Miller VL (2007) The RovA
regulons of Yersinia enterocolitica and Yersinia pestis are distinct: evidence that manyRovA-regulated genes were acquired more recently than the core genome. Mol
Microbiol 66: 189–205.92. Vizcaino JA, Cote RG, Csordas A, Dianes JA, Fabregat A, et al. (2013) The
PRoteomics IDEntifications (PRIDE) database and associated tools: status in
2013. Nucleic Acids Res 41: D1063–1069.93. Heesemann J, Laufs R (1983) Construction of a mobilizable Yersinia enterocolitica
virulence plasmid. J Bacteriol 155: 761–767.94. Antonenko V, Pawlow V, Heesemann J, Rakin A (2003) Characterization of a
novel unique restriction-modification system from Yersinia enterocolitica O:8 1B.FEMS Microbiol Lett 219: 249–252.
contribution of Yersinia enterocolitica virulence factors to evasion of microbicidalaction of neutrophils. Infect Immun 64: 724–733.
96. Pelludat C (2000) Molekulare Charakterisierung der Pathogenitatsinsel von
Yersinia enterocolitica und ihr Einfluss auf die aktive Eisenaufnahme. Wuerzburg:Julius-Maximilians-Universitaet Wuerzburg. 117 p.
97. Kinder SA, Badger JL, Bryant GO, Pepe JC, Miller VL (1993) Cloning of theYenI restriction endonuclease and methyltransferase from Yersinia enterocolitica
serotype O8 and construction of a transformable R-M+ mutant. Gene 136: 271–
275.98. Hanahan D (1983) Studies on transformation of Escherichia coli with plasmids. J
Mol Biol 166: 557–580.99. Herrero M, de Lorenzo V, Timmis KN (1990) Transposon vectors containing
non-antibiotic resistance selection markers for cloning and stable chromosomalinsertion of foreign genes in gram-negative bacteria. J Bacteriol 172: 6557–6567.
Yersinia enterocolitica RNA Chaperone Hfq
PLOS ONE | www.plosone.org 17 January 2014 | Volume 9 | Issue 1 | e86113